Predicting Cesarean Section Deliveries Using Hybrid Machine Learning Models: Evidence from the Bangladesh Demographic and Health Survey

Abstract

Cesarean section rates in Bangladesh have risen sharply over the past decade, raising concerns about over-medicalization. Using data from the Bangladesh Demographic and Health Survey (BDHS), we develop hybrid ML models integrating logistic regression, random forest, and XGBoost to predict C-section likelihood from sociodemographic and clinical features.

Citation

Conference: International Conference on Applied Statistics and Data Science (ICASDS-2025)
Status: Poster Accepted, Paper ID: 189